Research shows that it is often the people with strong opinions who leave reviews of products they’ve bought and service they’ve received and the majority without strong opinions tend not to leave so many reviews.

So what about that silent majority?

Research at MIT suggests that some reviewers may be systematically biased or easily manipulated by the presence of previous reviewers comments. This is commonly called “social proof” where people assume the actions of others are correct and duplicate them in an attempt to reflect correct behaviour in the situation.

Imagine you dine at a restaurant and aren’t too impressed by the food. When you get home and want to post a review, you find that the restaurant has very high ratings – people love the place. You may still post a negative review but it is quite likely you will assume your not so good dish was an exception for the restaurant and will hold off on posting that negative view or at least tone it down.

When we see that other people have appreciated a certain book, enjoyed a hotel or restaurant or liked a particular event — and given them a high rating online — this can cause us to feel the same positive feelings about the book, hotel, restaurant or event and to likewise provide a similarly high online rating,

If you had a moderate view on a restaurant meal or event etc. you’re likely not to bother leaving a review, thinking it not worth the time and effort.

An academic study titled “Understanding and Overcoming Biases in Customer Reviews.” had analysis of several hundred thousand reviews from four major online retailers, and highlighted evidence of two major types of bias in the online review system:  social influence bias and selection bias.

Social Influence Bias is when a user’s opinion is influenced by the opinion of others. So, if your business has bad reviews, people who post reviews are more likely to follow suit, and post more bad reviews. If your business has good reviews, people are more likely to post good reviews.

Selection Bias (also called voluntary response) is where the people that submit reviews feel motivated to do so, which usually means that the resulting sample over-represents individuals who have strong opinions.

There is also the problem of how to interpret rating scales. For example, does 5 out of 5 stars means exceptional or does it just mean very good?

A well designed mass survey will put considerable effort towards standardizing responses, as vague and inconsistent response criteria will make an otherwise legitimate survey meaningless.

It can be difficult to get wide ranging responses from  groups of people without ignoring the silent majority but use of incentives can make it possible.

Do you have an opinion on this matter? Please comment in the box below.